論文の内容の要旨
Investigation into New Catalytic Systems for Cross-Dehydrogenative
Coupling Reactions under Aerobic Conditions
（酸素雰囲気下における脱水素クロスカップリング反応のための新規
触媒系の研究）
田上
新
Introduction
Over the past hundred years, significant progress in organic chemistry has been made.
However, regio-selective C-H functionalization still remains a challenging topic. Recently,
cross-dehydrogenative coupling (CDC) reactions of amines or ethers have emerged as a novel
method in which -selective C-H functionalization is achievable. Thus far, several transition
metals have been reported as catalysts for CDC reactions. As for the metal-free reactions,
some methods have been developed but examples are still rare. In addition, these oxidative
coupling reactions often required stoichiometric amounts of oxidants such as t-BuOOH, DDQ,
and so on. Thus, the use of molecular oxygen as an oxidant is more beneficial for the
environment and a desired goal. Based on these facts, I decided to explore novel metal-free
CDC reactions of amines under aerobic conditions. During the course of my investigation, I
discovered antimony and sulfuryl chloride as novel catalysts for transition metal-free and
metal-free reactions, respectively.
1. Antimonate/NHPI-Catalyzed CDC Reaction of Tertiary Amines
In order to achieve a metal-free CDC reaction, I hypothesized that aminium radical cation
could serve as a catalyst. Aminium radicals are known as single electron oxidants, whose
oxidation potentials are tunable by modifying the substituents on aromatic moiety. Since
substoichiometric use of these species for an oxidation reaction was recently reported, there
was the possibility to catalyze CDC reactions. In addition, I hypothesized that Nhydroxyphthalimide (NHPI), which is effective for hydrogen radical abstraction, would
promote the reaction. Thus, I initially attempted the nitromethylation reaction of N-
phenyltetrahydroisoquinoline (1a) under
oxygen
atmosphere
using
tris(pbromophenyl)aminium
hexachloroantimonate and NHPI as catalysts (Table 1,
entry 1). Although antimony is classified as a
semi-metal, I speculated that counter anion
could be replaced at a latter stage. To our
delight, nitromethylated product (2a) was
furnished in a good yield. In order to survey
the
effect
of
counter
anions,
hexafluoroantimonate salt was prepared and
tested, which also provided the product in a
modest yield (entry 2). However, when tetrafluoroborate or hexafluorophosphate salts
were utilized, the reactions were very
sluggish (entries 3-4). Instead, when the
counter
cation
was
replaced
with
tetraethylammonium or trityl cations, 2a was obtained in modest yields (entries 5-6). Based
on these experiments, I concluded that antimonate anion served as a catalyst for the oxidative
coupling reaction. The use of antimony in the oxidative coupling reaction seemed interesting
since: (i) antimony is semi-metal and inexpensive; (ii) the use of antimony in CDC reaction of
amines has not been reported; and (iii) hexahaloantimonate anions have been generally
considered as stable and innocent counter anions in organic chemistry. Thus, I decided to
further investigate antimonate as a catalyst. Since nitromethylation was also catalyzed by
antimony pentachloride, I determined that antimony was the real active species (entry 7).
After examining various sources of antimony, sodium hexachloroantimonate turned out to be
the best catalyst (entry 8).
After optimization of the reaction conditions, the substrate generality of this oxidative
coupling reaction was investigated (Table 2). In addition to N-phenyl-substituted
tetrahydroisoquinolines, substrates
which possess electron-rich aromatic
groups furnished the products in
high yields (entries 1-5). On the
other hand, an electron-deficient
amine was less reactive (entry 6). In
addition to nitromethane, nitroethane
was also shown to react smoothly
despite with modest regio-selectivity
(entry 7). In addition to nitroalkanes,
various nucleophiles were applicable
under similar reaction conditions.
2. Zinc Antimonate-Catalyzed Oxidative Allylation of -Amino Acid Derivatives
Although
antimonycatalyzed CDC reaction of
tertiary
amines
was
attractive, viable substrates
were limited only to Naryltetrahydroisoquinolines.
In order to solve this
problem, I hypothesized that N-aryl-substituted glycine derivatives would be alternative
candidates since higher acidity of -hydrogen of -amino acid derivatives may facilitate
oxidation process. In addition to this hypothesis, I developed a novel concept of employing
metal antimonate as a new class of bifunctional catalyst, where the metal cation serves as a
Lewis acid to promote nucleophilic addition and the antimonate anion serves as a catalyst for
aerobic oxidation. Based on these hypotheses, I initially attempted the CDC reactions using
protic nucleophiles such as nitromethane. However, none of these nucleophiles furnished
desired products in acceptable yields. Instead, it turned out that allyltributyltin was a stable
and reactive nucleophile. Among several metal Lewis acids I examined, zinc(II) was found to
be the best Lewis acid. Thus, zinc hexachloroantimonate was prepared and employed as a
bifunctional catalyst. Although zinc antimonate served as a bifunctional catalyst for the
oxidative allylation reaction of glycine derivatives, several disadvantages limited the synthetic
utility of this catalyst. For example, since oxidation was inhibited in the presence of allyltin,
allyltin was added after completion of oxidation reaction. In addition, it was found that not
only zinc, but also antimony species seemed to serve as a Lewis acid. Finally, due to
competing side-reactions, yields were relatively moderate.
After optimization of reaction
conditions, the substrate scope of
this oxidative allylation reaction
was investigated (Table 3). In
addition to a glycine ethyl ester,
which has p-methoxyphenyl (PMP)
group on the nitrogen, methyl,
isopropyl, benzyl esters provided
the allylated products in good yields
(entries 1, 3, 5), while sterically
hindered t-butyl ester was less
reactive (entry 4). As for glycine
amides, the methyl amide gave the
corresponding product in a high
yield (entry 6), while ethyl and
benzyl amides afforded products in
modest yields (entries 7, 8). Pyrrolidine-substituted amide was also found to be reactive (entry
9). When the aryl group on the nitrogen was modified, 2,6-dimethyl-4-methoxyphenyl
substrate reacted smoothly (entry 10). On the other hand, no oxidation took place for the 2,6dibromo-4-methoxyphenyl substrate (entry 11). Compared with the PMP substrate, p-tolyl
and phenyl substrates were less reactive (entries 12, 13). Finally, when a peptide was
subjected to the same reaction conditions, the -hydrogen of the glycine which possess PMP
group on it was selectively functionalized (entry 14).
3. Sulfuryl Chloride-mediated CDC reaction of Tertiary N-Aryl Amines
During the course of my investigation into the antimony-catalyzed oxidative coupling
reactions, it was found that N-chlorosuccinimide (NCS) can serve as a catalyst for the CDC
reaction of N-phenyltetrahydroisoquinoline (Table 4, entry 1). When other electrophilic
chlorine sources were examined, most
catalysts provided the desired product in
low yields (entries 2-5). On the other hand,
sulfuryl chloride turned out to be the most
effective catalyst (entry 6). Further
optimization revealed that the use of 6
mol% sulfuryl chloride afforded the
product in an excellent yield (entry 7). In
contrast, reactions with other electrophilic
halogen
sources
such
as
Nbromosuccinimide (NBS) and iodine were
very sluggish (entries 8, 9). The reactions
are believed to proceed through a radicalinitiated autoxidation mechanism.
Conclusion
Antimony was found to be a novel catalyst for the cross-dehydrogenative coupling reactions
of tertiary amines under mild aerobic conditions in the presence of NHPI as a co-catalyst.
Based on this initial discovery, a novel bifunctional catalyst was developed for the functionalization of more synthetically useful glycine derivatives. Although antimony is
classified as a semi-metal, the use of antimony for CDC reactions is novel since
hexachloroantimonate anion has long been recognized as a stable and innocent counter anion
in the field of organic chemistry. The present work dispels this widely regarded assumption.
Finally, the metal-free CDC reaction was achieved by using sulfuryl chloride as a non-metal
catalyst.